CN116886591B - Computer network system and routing method - Google Patents
Computer network system and routing method Download PDFInfo
- Publication number
- CN116886591B CN116886591B CN202311139923.0A CN202311139923A CN116886591B CN 116886591 B CN116886591 B CN 116886591B CN 202311139923 A CN202311139923 A CN 202311139923A CN 116886591 B CN116886591 B CN 116886591B
- Authority
- CN
- China
- Prior art keywords
- ring unit
- node
- units
- network nodes
- network
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 21
- 230000005540 biological transmission Effects 0.000 claims abstract description 62
- 238000004891 communication Methods 0.000 claims abstract description 17
- 230000008569 process Effects 0.000 abstract description 8
- 230000000694 effects Effects 0.000 abstract 1
- ODCKICSDIPVTRM-UHFFFAOYSA-N [4-[2-hydroxy-3-(propan-2-ylazaniumyl)propoxy]naphthalen-1-yl] sulfate Chemical compound C1=CC=C2C(OCC(O)CNC(C)C)=CC=C(OS(O)(=O)=O)C2=C1 ODCKICSDIPVTRM-UHFFFAOYSA-N 0.000 description 21
- 235000008694 Humulus lupulus Nutrition 0.000 description 9
- 125000004122 cyclic group Chemical group 0.000 description 4
- 238000012986 modification Methods 0.000 description 4
- 230000004048 modification Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000004904 shortening Methods 0.000 description 3
- 230000004075 alteration Effects 0.000 description 2
- 238000005516 engineering process Methods 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 238000013461 design Methods 0.000 description 1
- 238000005265 energy consumption Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000002093 peripheral effect Effects 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/12—Shortest path evaluation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/42—Loop networks
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L45/00—Routing or path finding of packets in data switching networks
- H04L45/02—Topology update or discovery
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L49/00—Packet switching elements
- H04L49/10—Packet switching elements characterised by the switching fabric construction
- H04L49/109—Integrated on microchip, e.g. switch-on-chip
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L12/00—Data switching networks
- H04L12/28—Data switching networks characterised by path configuration, e.g. LAN [Local Area Networks] or WAN [Wide Area Networks]
- H04L12/42—Loop networks
- H04L2012/421—Interconnected ring systems
Landscapes
- Engineering & Computer Science (AREA)
- Computer Networks & Wireless Communication (AREA)
- Signal Processing (AREA)
- Data Exchanges In Wide-Area Networks (AREA)
Abstract
The embodiment of the application provides a topology structure and a routing method of a network on chip, which relate to the technical field of electronic physics and comprise the following steps: each annular unit comprises at least three network nodes, communication connection is established between two adjacent annular units in the at least three annular units, and the network nodes included in each annular unit are symmetrically distributed; establishing communication connection between two adjacent network nodes; each network node is used for transmitting data when serving as a source node; and/or for receiving data when acting as a destination node; and/or for forwarding data when acting as an intermediate node. The technical problem that the transmission delay is high due to the existing network structure is solved, the number of nodes passing through in the middle of the process of calculating network transmission is reduced, the data transmission time is shortened, and the technical effect of reducing the transmission delay of data routing is achieved.
Description
Technical Field
The present application relates to the field of electronic physical technology, and in particular, to a computer network system and a routing method.
Background
Along with the increasing computational complexity of communication terminal devices, the integration scale of real-time complex system chips is gradually increased, and Network on chips (Network on chips) are also generated, wherein the topology structure of the Network on Chip defines the physical layout of each module in the Network for distribution and connection on the Chip. The selection of the topology structure directly affects the network node degree, the network diameter and the network scale, thereby affecting the network delay, the throughput, the energy consumption, the area, the fault tolerance and the like, and finally having important influence on the network performance parameters. Therefore, in the network on chip, design research on the topology structure is particularly important.
At present, specific topological structures such as Ring, network Mesh, torus, hierarchical Ring and the like can be adopted in the related technology, however, more nodes need to be passed through for transmission delay in the structure, and the transmission delay is high.
Disclosure of Invention
The embodiment of the application provides a topology structure of a network on chip and a routing method.
In a first aspect of an embodiment of the present application, there is provided a topology of a network on chip, the topology including: each annular unit comprises at least three network nodes, communication connection is established between two adjacent annular units in the at least three annular ternary units, and the network nodes included in each annular unit are symmetrically distributed; establishing communication connection between two adjacent network nodes;
each network node is used for transmitting data when serving as a source node; and/or the number of the groups of groups,
for receiving data when acting as a destination node; and/or the number of the groups of groups,
for forwarding data when acting as an intermediate node.
In an alternative embodiment of the present application, the at least three cyclic units are distributed in a cyclic manner from inside to outside.
In an alternative embodiment of the present application, the at least three ring units include an outer ring unit, an intermediate ring unit and an inner ring unit, the outer ring unit being located at an outermost layer of the topology, the inner ring unit being located at an innermost layer of the topology, the intermediate ring unit being located at an intermediate layer of the topology;
the network node identifier contained in the outer annular unit is larger than the network node identifier contained in the middle annular unit; the network node identifier contained in the intermediate ring unit is larger than the network node identifier contained in the inner ring unit.
In an alternative embodiment of the present application, the number of network nodes included in the outer ring unit is greater than or equal to the number of network nodes included in the inner ring unit.
In an alternative embodiment of the present application, in the adjacent three ring units of the topology structure, the number of network nodes included in the ring unit located at the inner layer and the ring unit located at the outer layer is greater than the number of network nodes included in the ring unit located at the middle layer.
In an alternative embodiment of the present application, when the adjacent three ring units are an outer ring unit, an intermediate ring unit, and an inner ring unit, the number of network nodes included in the outer ring unit and the inner ring unit is greater than the number of network nodes included in the intermediate ring unit.
In an optional embodiment of the present application, when the adjacent three ring units are adjacent three middle ring units sequentially arranged from inside to outside, the number of network nodes included in the middle ring unit located in the inner layer and the number of network nodes included in the middle ring unit located in the outer layer are both greater than the number of network nodes included in the middle ring unit located in the middle layer.
In an optional embodiment of the present application, when the adjacent three ring units are two adjacent middle ring units and outer ring units sequentially arranged from inside to outside, the number of network nodes included in the middle ring unit and the outer ring unit located in the inner layer is greater than the number of network nodes included in the middle ring unit located in the outer layer.
In an optional embodiment of the present application, when the adjacent three ring units are an inner ring unit and two adjacent middle ring units, which are sequentially arranged from inside to outside, the number of network nodes contained in the inner ring unit and the number of network nodes contained in the middle ring unit located at the outer layer are both greater than the number of network nodes contained in the middle ring unit located at the inner layer.
In a second aspect of the embodiments of the present application, a routing method is provided, applied to the topology structure as set forth in the foregoing claims, where the method includes:
receiving and responding to a route transmission request, and acquiring a source node and a destination node in the topological structure;
determining an optimal transmission path from the source node to the destination node based on the topology; the transmission delay of the optimal transmission path is minimum;
and transmitting data included in the route transmission request from the source node to the destination node based on the optimal transmission path.
The topology structure of the network on chip provided by the embodiment of the application comprises at least three ring units, wherein each ring unit comprises at least three network nodes, communication connection is established between two adjacent ring units in at least three ring ternary, and the network nodes included in each ring unit are symmetrically distributed; establishing communication connection between two adjacent network nodes; each network node is used for transmitting data when serving as a source node; and/or for receiving data when acting as a destination node; and/or for forwarding data when acting as an intermediate node. Compared with the prior art, the technical scheme of the application has the advantages that at least three annular units are arranged, communication connection is established between two adjacent annular units in the at least three annular units, the network topology structure is greatly simplified, and the number of nodes passing through in the middle of the process of calculating network transmission is reduced due to the symmetrical distribution of network nodes included in each annular unit, so that the data transmission time is shortened, and the transmission delay of data routing is greatly reduced.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this specification, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute a limitation on the application. In the drawings:
fig. 1 is a schematic structural diagram of a conventional mesh network structure according to an embodiment of the present application;
FIG. 2 is a schematic diagram of a topology according to an embodiment of the present application;
FIG. 3 is a schematic diagram of a topology according to another embodiment of the present application;
fig. 4 is a flowchart of a routing method according to an embodiment of the present application.
Detailed Description
In carrying out the present application, the inventors have found that the average latency of the topology of current network-on-chip is high.
For example, referring to fig. 1, a mesh network structure is taken as an example, where nodes are arranged to form a regular grid, each node is only connected to neighboring nodes in the same row and column, the network structure includes 16 nodes, which are nodes 1, 2, 3, 4, 15, 16, etc., each node is connected to form a network interface, and the 16 nodes are arranged to form a 4×4 network. In the mesh network structure, each circle represents a node, and each node can be used as a source node to send data, can be used as a target node to receive data, and can be used as an intermediate node to forward data. Wherein the data path between adjacent nodes is bi-directional from the current node to the adjacent nodes may be referred to as a one-hop (1 hop).
In the mesh network structure, the transmission delay between the source node and the destination node can be calculated, and since the nodes 1, 2, 5, 6 and the nodes 3, 4, 7, 8 and the nodes 9, 10, 13, 14 and the nodes 11, 12, 15, 16 are symmetrical, only the average delay from the nodes 1, 2, 5, 6 to other nodes needs to be calculated when the average delay of the whole mesh network structure is calculated. The total delay between the node 1 and other nodes is respectively: 1- >2:1hop;1- >3:2hops;1- > 4:3hops; 1- >5:1hop;1- >6:2hops;1- >7:3hops;1- > 8:4 hops;1- >9:2hops;1- >10:3hops;1- >11:4hops;1- > 12:5 hops;1- >13:3hops;1- >14:4hops;1- >15:5hops;1- > 16:6 hops, the total delay of node 1 and other nodes is 48hops.
Accordingly, the total delay between node 2 and the other nodes is: 2- > 1:1 hop;2- > 3:1 hop;2- >4:2hops; 2- > 5:2hops; 2- > 6:1hop; 2- >7:2hops; 2- > 8:3 hops;2- > 9:3 hops;2- >10:2hops;2- >11:3hops;2- >12:4hops;2- >13:4hops;2- >14:3hops;2- >15:4hops;2- >16:5hops; the total delay of node 2 with the other nodes is 40hops.
The total delay between node 5 and the other nodes is: 5- > 1:1 hop;5- > 2:2hops; 3:3hops;5- > 4:4hops; 5- > 6:1hop; 5- >7:2hops;5- > 8:3 hops;5- >9:1hop;5- >10:2hops;5- >11:3hops;5- >12:4hops;5- >13:2hops;5- >14:3hops;5- >15:4hops;5- >16:5hops; the total delay of node 5 with the other nodes is 40hops.
The total delay between node 6 and the other nodes is: 6- > 1:2hops;6- >2:1hop; 6- >3:2hops; 6- > 4:3hops; 6- >5:1hop; 6- > 7:1 hop;6- >8:2hops; 6- >9:2hops; 6- >10:1hop;6- >11:2hops;6- >12:3hops;6- >13:3hops;6- >14:2hops;6- >15:3hops;6- >16:4hops; the total delay between node 6 and the other nodes is 32hops.
Then according to the total delay of 1, 2, 5 and 6 nodes, the average delay of the mesh network structure is calculated as = (48+40+40+32)/415=2.67hops。
In view of the above problems, the embodiments of the present application provide a topology structure and a routing method to reduce transmission delay of data routing.
The schemes in the embodiments of the present application may be implemented in various computer languages, for example, in various hardware description languages, such as Verilog, systemVerilog, chisel, systemC, etc.
In order to make the technical solutions and advantages of the embodiments of the present application more apparent, the following detailed description of exemplary embodiments of the present application is provided in conjunction with the accompanying drawings, and it is apparent that the described embodiments are only some embodiments of the present application and not exhaustive of all embodiments. It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other.
Referring to fig. 2, the topology structure of the network on chip includes at least three ring units, each ring unit includes at least three network nodes, communication connection is established between two adjacent ring units in the at least three ring units, and the network nodes included in each ring unit are symmetrically distributed; a communication connection is established between two adjacent network nodes. Each network node is used for transmitting data when serving as a source node; and/or for receiving data when acting as a destination node; and/or for forwarding data when acting as an intermediate node.
Specifically, the topology structure includes at least three ring units, each ring unit is distributed in a ring shape, each ring unit includes at least three network nodes, and each network node may be a source node, a destination node, or an intermediate node.
Each network node in the ring unit forms a closed ring, each node in the ring unit is connected in a closed ring communication line connected end to end through a loop interface, and any node in the ring unit can request to send data, receive data or forward data to the next node.
Alternatively, each network node within the topology may be a CPU, DMA, DDR, subsystem, or even other peripheral. The embodiment of the present application is not particularly limited thereto.
The two adjacent ring units can be connected through one network node, can be connected through two network nodes, and can be connected through three or more network nodes. Wherein each network node connected is located on two adjacent ring units. The individual ring units described above may be distributed in a nested configuration.
Referring to fig. 3, the topology of 16 nodes, namely nodes 1, 2, 3, 4, 15, 16, etc., are shown, each node is connected to form a network interface, the node comprises four ring units, for example, nodes 1, 2, 3, 4 form a ring unit, nodes 5, 6, 7, 8 form a ring unit, nodes 9, 10, 11, 12 form a ring unit, and nodes 13, 14, 15, 16 form a ring unit.
For this multi-ring nested topology, since nodes 1, 5, 9, 13 and nodes 2, 6, 10, 14 and nodes 4, 8, 12, 16 and nodes 3, 7, 11, 15 are symmetrical, the average delay of the whole structure can be obtained by calculating the average delay of nodes 1, 5, 9, 13.
The total delay between the node 1 and other nodes is respectively: 1- >2:1hop;1- >3:2hops;1- >4:1hop;1- >5:1hop;1- >6:2hops;1- >7:3hops;1- >8:2hops;1- >9:2hops;1- >10:3hops;1- >11:4hops;1- >12:3hops;1- >13:3hops;1- >14:4hops;1- >15:5hops;1- >16:4hops; the total delay of node 1 with the other nodes is 40hops.
Accordingly, the total delay between node 5 and the other nodes is: 5- > 1:1 hop;5- > 2:2hops;5- >3:3hops;5- >4:2hops;5- > 6:1hop; 5- >7:2hops;5- >8:1hop;5- >9:1hop;5- >10:2hops;5- >11:3hops;5- >12:2hops;5- >13:2hops;5- >14:3hops;5- >15:4hops;5- >16:3hops; the total delay of node 5 with other nodes is 32hops.
The total delay between node 9 and the other nodes is: 9- > 1:2hops;9- > 2:3hops; 9- > 3:4hops; 9- > 4:3hops; 9- >5:1hop; 9- >6:2hops; 9- >7:3hops;9- >8:2hops;9- >10:1hop;9- >11:2hops;9- >12:1hops;9- >13:1hops;9- >14:2hops;9- >15:3hops;9- >16:2hops; the total delay of node 9 with the other nodes is 32hops.
The total delay between node 13 and the other nodes is: 13- > 1:3hops;13- > 2:4hops; 13- > 3:5 hops;13- > 4:4hops; 13- > 5:2hops;13- > 6:3hops; 13- > 7:4 hops;13- > 8:3 hops;13- >9:1 hops;13- >10:2hops;13- >11:3hops;13- >12:2hops;13- >14:1hop;13- >15:2hops;13- >16:1hop; the total delay of node 13 with the other nodes is 40hops.
Then, according to the total delay of 1, 5, 9 and 13 nodes, the average delay of the topological structure is calculated as = (40+32+32+40)/415 = 2.4hops。
When the network nodes are 16 nodes, compared with the existing mesh network structure, the average delay of the topology structure provided by the application is 2.4hops less than that of the mesh network structure, and the average delay is reduced by 2.67-2.4=0.23 hops, and the reduction ratio is about 10%, so that the average delay is greatly shortened.
The topology structure of the network on chip provided by the embodiment of the application comprises at least three ring units, wherein each ring unit comprises at least three network nodes, communication connection is established between two adjacent ring units in the at least three ring units, and the network nodes included in each ring unit are symmetrically distributed; establishing communication connection between two adjacent network nodes; each network node is used for transmitting data when serving as a source node; and/or for receiving data when acting as a destination node; and/or for forwarding data when acting as an intermediate node. Compared with the prior art, the technical scheme of the application has the advantages that at least three annular units are arranged, communication connection is established between two adjacent annular units in the at least three annular units, the network topology structure is greatly simplified, and the number of nodes passing through in the middle in the process of calculating network transmission is reduced due to the symmetrical distribution of network nodes included in each annular unit, so that the data transmission time is shortened, and the transmission delay of data routing is greatly reduced.
In an alternative embodiment of the present application, the at least three cyclic units are distributed in a cyclic manner from inside to outside.
Specifically, the annular distribution of each annular unit from inside to outside may mean that the network node identifiers sequentially extend from small to large to outside, and the network nodes of each annular unit are arranged in the same order.
For example, when the topology includes three ring units, the network node identifiers in each of the three ring units are arranged in a clockwise direction and in a closed manner, or are arranged in a counterclockwise direction and in a closed manner. Each ring unit may include three network nodes, or may include four network nodes or even more network nodes.
In the embodiment, at least three annular units are annularly distributed from inside to outside, so that the topological structure can be simplified, and the data transmission delay is greatly shortened.
In an alternative embodiment of the present application, the at least three ring units include an outer ring unit, an intermediate ring unit, and an inner ring unit, the outer ring unit is located at an outermost layer of the topology, the inner ring unit is located at an innermost layer of the topology, and the intermediate ring unit is located at an intermediate layer of the topology.
The network node identifier contained in the outer annular unit is larger than the network node identifier contained in the middle annular unit; the intermediate ring unit contains a network node identifier greater than the inner ring unit.
With continued reference to fig. 2, the topology structure includes four ring units, where the outer ring unit includes four nodes, namely nodes 13, 14, 15 and 16, the middle ring unit includes two nodes, each middle ring unit includes different node identifiers, and the node identifier included in the middle ring unit located at the outer layer is greater than the node identifier included in the middle ring unit located at the inner layer, that is, the node identifiers included in the middle ring unit located at the outer layer are respectively 9, 10, 11 and 12, and the node identifiers included in the middle ring unit located at the inner layer are respectively 5, 6, 7 and 8. The inner layer ring unit contains 4 network nodes, and the corresponding node identifiers are 1, 2, 3 and 4 respectively.
Wherein each annular unit is distributed from inside to outside according to the sequence from the small mark to the large mark. And establishing communication connection between the outer annular unit, the middle annular unit and the network nodes contained in the inner annular unit in pairs.
In the embodiment, the outer annular unit, the middle annular unit and the inner annular unit are arranged, and the network node identifier contained in the outer annular unit is larger than the network node identifier contained in the middle annular unit; the network node identifiers contained in the middle annular unit are larger than those contained in the inner annular unit, so that the network structure of the topological structure is greatly simplified, the number of intermediate nodes passing through the topological structure in the data transmission process is small, and the data transmission delay can be ensured to be smaller.
In an alternative embodiment of the application, the number of network nodes comprised in the outer ring unit is greater than or equal to the number of network nodes comprised in the inner ring unit.
When the number of network nodes is large, the number of network nodes included in the outer ring unit may be greater than or equal to the number of network nodes included in the inner ring unit. For example, when the number of network nodes included in the outer ring unit may be four, the number of network nodes included in the inner ring unit may also be four. When the number of network nodes included in the outer ring unit may be sixteen, the number of network nodes included in the inner ring unit may be four.
According to the embodiment of the application, the number of the network nodes included in the outer layer annular units is larger than or equal to that of the network nodes included in the inner layer annular units, so that the annular units with the number of multiple nodes can be distributed on the outer layer of the topological structure, the topological structure is ensured to be more regular, and the network topological structure is simplified.
In an alternative embodiment of the present application, in the adjacent three ring units sequentially arranged from inside to outside in the topology structure, the number of network nodes included in the ring unit located at the inner layer and the ring unit located at the outer layer is greater than the number of network nodes included in the ring unit located at the middle layer.
It should be noted that, as the number of nodes in a single ring unit increases, the average delay increases, and to some extent, the delay may be larger than that of the mesh network. In this case, a ring unit having a relatively small number of nodes may be further nested between two ring units having a relatively large number of nodes.
Referring to fig. 3, the topology may include three ring units including network nodes 1, 2, 3, 4, & gt, 42, 43, 44, wherein, among the adjacent three ring units sequentially arranged from inside to outside, the ring units located at the inner layer include network nodes 1, 2, 3, 4, respectively, the ring units located at the middle layer include network nodes 13, 14, 15, and 16, and the ring units located at the outermost layer include network nodes 17, 18, 19.
For the above network topology, the shortest path is, for example, 38-10-9-8-7-6-5-4-24 when the source node is the 38 # node, the destination node is the 24 # node, and the total transmission delay is 8hops when there is no intermediate ring unit. By adopting the topology structure of the application, the intermediate ring units exist, the number of network nodes included in the intermediate ring units is not more than the number of network nodes included in the ring units positioned at the inner layer and the ring units positioned at the outer layer, the shortest path can be determined to be 38-16-15-14-24, the total transmission delay is 4hops, and obviously, the topology structure of the application can reduce the transmission delay.
In this embodiment, the number of network nodes included in the ring units located at the inner layer and the ring units located at the outer layer is greater than the number of network nodes included in the ring units located at the middle layer, so that fewer intermediate nodes pass through in the process of calculating the transmission delay, the determined shortest path is better, and the network transmission delay is reduced.
In an alternative embodiment of the present application, when the adjacent three ring units are an outer ring unit, an intermediate ring unit and an inner ring unit, the number of network nodes included in the outer ring unit and the inner ring unit is greater than the number of network nodes included in the intermediate ring unit.
Specifically, when the number of the intermediate ring units in the topology structure is one, one intermediate ring unit with a relatively small number of nodes can be nested between the two outer ring units and the inner ring unit with a relatively large number of nodes. For example, if the number of nodes included in the outer ring unit is 20 and the number of nodes included in the inner ring unit is 10, the number of nodes included in the middle ring unit may be 4.
The number of the ring units and the number of the network nodes of each ring unit can be determined according to the number of the network nodes in an actual scene, so that a lower transmission delay is achieved when data is transmitted between a source node and a destination node.
Specifically, when the source node is located in the outer ring unit, the destination node is located in the inner ring unit, or the destination node is located in the outer ring unit, and when the source node is located in the inner ring unit, the number of network nodes included in the outer ring unit and the inner ring unit is greater than that of network nodes included in the intermediate ring unit, then in the process of calculating the transmission delay between the source node and the destination node, the shortest path may be determined first, and the shortest path may pass through the nodes in the intermediate ring unit with a smaller number, that is, the shortest path is the source node-the node of the intermediate ring unit-the destination node, so that the transmission delay between the source node and the destination node is smaller.
In this embodiment, the number of network nodes included in the outer ring unit and the inner ring unit is greater than the number of network nodes included in the intermediate ring unit, so that when the source node and the destination node are respectively located in the outer ring unit and the inner ring unit, the shortest path is determined by the nodes in the intermediate ring unit, thereby shortening the data transmission delay.
In an alternative embodiment of the present application, when the adjacent three ring units are adjacent three middle ring units sequentially arranged from inside to outside, the number of network nodes included in the middle ring unit located at the inner layer and the number of network nodes included in the middle ring unit located at the outer layer are both greater than the number of network nodes included in the middle ring unit located at the middle layer.
Specifically, when there are a plurality of intermediate ring units in the topology structure, an intermediate ring unit with a relatively small number of nodes may be nested between the intermediate ring unit located in the inner layer and the intermediate ring unit located in the outer layer, where the intermediate ring unit with a relatively small number of nodes is the intermediate ring unit located in the intermediate layer. For example, the number of nodes included in the intermediate ring unit located at the outer layer is 20, and the number of nodes included in the intermediate ring unit located at the inner layer is 10, and the number of nodes included in the intermediate ring unit located at the intermediate layer may be 4.
When the source node is located in the middle ring unit of the outer layer, the destination node is located in the middle ring unit of the inner layer, or the destination node is located in the middle ring unit of the outer layer, if the number of network nodes included in the middle ring unit of the outer layer and the middle ring unit of the inner layer is greater than the number of network nodes included in the middle ring unit of the middle layer, the shortest path can be determined first in the process of calculating the transmission delay between the source node and the destination node, and the shortest path can pass through a smaller number of nodes located in the middle ring unit of the middle layer, namely, the shortest path is the node-destination node of the source node-middle ring unit, so that the transmission delay between the source node and the destination node is smaller.
In this embodiment, the number of network nodes included in the intermediate ring unit located at the inner layer and the intermediate ring unit located at the outer layer is greater than the number of network nodes included in the intermediate ring unit located at the middle layer, so that when the source node and the destination node are located at the outer layer and the intermediate ring unit located at the inner layer respectively, the shortest path is determined by the nodes in the intermediate ring unit located at the middle layer, thereby shortening the data transmission delay.
In an alternative embodiment of the present application, when the adjacent three ring units are two adjacent middle ring units and outer ring units sequentially arranged from inside to outside, the number of network nodes included in the middle ring unit and the outer ring unit located in the inner layer is greater than the number of network nodes included in the middle ring unit located in the outer layer.
Specifically, when there are a plurality of intermediate ring units in the topology structure, an intermediate ring unit with a relatively small number of nodes may be nested between the intermediate ring unit located in the inner layer and the outer ring unit, and the intermediate ring unit with a relatively small number of nodes is the intermediate ring unit located in the outer layer. For example, the number of nodes included in the outer ring unit is 20, and the number of nodes included in the middle ring unit located in the inner layer is 10, and the number of nodes included in the middle ring unit located in the outer layer may be 4.
When the source node is located in the outer ring unit, the destination node is located in the middle ring unit located in the inner layer, or the destination node is located in the outer ring unit, and the source node is located in the middle ring unit located in the inner layer, the number of network nodes included in the outer ring unit and the middle ring unit located in the inner layer is greater than the number of network nodes included in the middle ring unit located in the outer layer, so that in the process of calculating the transmission delay between the source node and the destination node, the shortest path can be determined first, and the shortest path can pass through the nodes in the middle ring unit located in the outer layer with a smaller number, namely, the shortest path is the node of the source node-the middle ring unit-the destination node, so that the transmission delay between the source node and the destination node is smaller.
In this embodiment, the number of network nodes included in the intermediate ring unit and the outer ring unit located at the inner layer is greater than the number of network nodes included in the intermediate ring unit located at the outer layer, so that when the source node and the destination node are located at the outer ring unit and the intermediate ring unit located at the inner layer respectively, the shortest path is determined by the nodes in the intermediate ring unit located at the outer layer, thereby shortening the data transmission delay.
In an alternative embodiment of the present application, when the adjacent three ring units are an inner ring unit and two adjacent intermediate ring units, which are sequentially arranged from inside to outside, the number of network nodes contained in the inner ring unit and the number of network nodes contained in the intermediate ring unit located on the outer layer are both greater than the number of network nodes contained in the intermediate ring unit located on the inner layer.
Specifically, when there are a plurality of intermediate ring units in the topology structure, an intermediate ring unit with a relatively small number of nodes may be nested between the inner ring unit and the intermediate ring unit located on the outer layer, where the intermediate ring unit with a relatively small number of nodes is the intermediate ring unit located on the inner layer. For example, the number of nodes included in the middle ring unit located at the outer layer is 20, and the number of nodes included in the inner ring unit is 10, and the number of nodes included in the middle ring unit located at the inner layer may be 4.
When the source node is located in the inner ring unit, the destination node is located in the outer middle ring unit, or the destination node is located in the inner ring unit, and the source node is located in the outer middle ring unit, the number of network nodes included in the inner ring unit and the outer middle ring unit is greater than the number of network nodes included in the inner middle ring unit, so that in the process of calculating the transmission delay between the source node and the destination node, the shortest path can be determined first, and the shortest path can pass through a smaller number of nodes located in the inner middle ring unit, namely, the shortest path is the node of the source node-the middle ring unit-the destination node, so that the transmission delay between the source node and the destination node is smaller.
In this embodiment, the number of network nodes included in the inner ring unit and the intermediate ring unit located on the outer layer is greater than the number of network nodes included in the intermediate ring unit located on the inner layer. When the source node and the destination node are respectively positioned in the inner annular unit and the middle annular unit positioned on the outer layer, the shortest path is determined through the nodes positioned in the middle annular unit positioned on the inner layer, and the data transmission delay is shortened.
On the other hand, in another embodiment of the present application, there is further provided a routing method for controlling the topology of the network on chip provided in the foregoing embodiment, as shown in fig. 4, where the routing method includes:
step 201, a source node and a destination node in a topology structure are acquired by receiving and responding to a route transmission request.
Step 202, determining an optimal transmission path from a source node to a destination node based on a topological structure; the transmission delay of the optimal transmission path is minimal.
Step 203, transmitting data included in the route transmission request from the source node to the destination node based on the optimal transmission path.
Specifically, when a Processing Element (PE) needs to transmit data from a source node to a destination node, the processing element packages the transmitted data, sends a routing request to a router according to position information of the destination node, and after receiving the routing request, the router responds to the routing request and routes the packaged data to other nodes, and the other nodes forward the data packet until the data packet reaches the destination node.
The source node and the destination node may be any two nodes in the topology.
After the source node and the destination node in the topological structure are acquired, an optimal transmission path between the source node and the destination node is determined based on the topological structure, when the source node and the destination node are located in two ring units with the number of network nodes being larger than a preset threshold value, whether one ring unit with the number not larger than the preset threshold value exists in the two ring units can be checked, and when the ring unit exists, the shortest path can pass through the nodes in the ring units with the smaller number to obtain the node-destination node of the source node-ring unit, so that the transmission delay between the source node and the destination node is smaller.
After the optimal transmission path is determined, the data is transmitted to the nodes in the annular units with the smaller number through the source node and then to the destination node, so that the data transmission is completed.
In the embodiment, the optimal transmission path between the source node and the destination node is determined based on the topological structure, and the data included in the route transmission request is transmitted from the source node to the destination node based on the optimal transmission path, so that the data transmission delay is greatly shortened, and the data transmission efficiency is improved.
It should be understood that, although the steps in the flowchart are shown in sequence as indicated by the arrows, the steps are not necessarily performed in sequence as indicated by the arrows. The steps are not strictly limited to the order of execution unless explicitly recited herein, and the steps may be executed in other orders. Moreover, at least some of the steps in the figures may include multiple sub-steps or stages that are not necessarily performed at the same time, but may be performed at different times, nor does the order in which the sub-steps or stages are performed necessarily performed in sequence, but may be performed alternately or alternately with at least a portion of other steps or other steps.
While preferred embodiments of the present application have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. It is therefore intended that the following claims be interpreted as including the preferred embodiments and all such alterations and modifications as fall within the scope of the application.
It will be apparent to those skilled in the art that various modifications and variations can be made to the present application without departing from the spirit or scope of the application. Thus, it is intended that the present application also include such modifications and alterations insofar as they come within the scope of the appended claims or the equivalents thereof.
Claims (9)
1. A computer network system, comprising: each annular unit comprises at least three network nodes, communication connection is established between two adjacent annular units in the at least three annular units, and the network nodes included in each annular unit are symmetrically distributed; establishing communication connection between two adjacent network nodes; in the adjacent three annular units which are sequentially arranged from inside to outside of the computer network system, the number of network nodes contained in the annular units positioned at the inner layer and the annular units positioned at the outer layer is larger than that of network nodes contained in the annular units positioned at the middle layer;
each network node is used for transmitting data when serving as a source node; and/or the number of the groups of groups,
for receiving data when acting as a destination node; and/or the number of the groups of groups,
for forwarding data when acting as an intermediate node.
2. The computer network system of claim 1, wherein the at least three ring units are distributed in a ring from inside to outside.
3. The computer network system of claim 2, wherein the at least three ring units include an outer ring unit, an intermediate ring unit, and an inner ring unit, the outer ring unit being located at an outermost layer of the computer network system, the inner ring unit being located at an innermost layer of the computer network system, the intermediate ring unit being located at an intermediate layer of the computer network system;
the network node identifier contained in the outer annular unit is larger than the network node identifier contained in the middle annular unit; the network node identifier contained in the intermediate ring unit is larger than the network node identifier contained in the inner ring unit.
4. A computer network system according to claim 3, wherein the number of network nodes included in the outer ring unit is greater than or equal to the number of network nodes included in the inner ring unit.
5. The computer network system of claim 1, wherein when the adjacent three ring units are an outer ring unit, an intermediate ring unit, and an inner ring unit, the number of network nodes included in both the outer ring unit and the inner ring unit is greater than the number of network nodes included in the intermediate ring unit.
6. The computer network system according to claim 1, wherein when the adjacent three ring units are adjacent three intermediate ring units arranged in order from inside to outside, the number of network nodes included in the intermediate ring unit located in the inner layer and the intermediate ring unit located in the outer layer is greater than the number of network nodes included in the intermediate ring unit located in the intermediate layer.
7. The computer network system according to claim 1, wherein when the adjacent three ring units are adjacent two intermediate ring units and outer ring units arranged in order from inside to outside, the number of network nodes included in the intermediate ring unit located in the inner layer and the number of network nodes included in the outer ring unit are each greater than the number of network nodes included in the intermediate ring unit located in the outer layer.
8. The computer network system according to claim 1, wherein when the adjacent three ring units are an inner ring unit and two adjacent intermediate ring units which are sequentially arranged from inside to outside, the inner ring unit and the intermediate ring unit located at the outer layer each have a number of network nodes greater than a number of network nodes included in the intermediate ring unit located at the inner layer.
9. A routing method applied to a computer network system according to any of the preceding claims 1-8, characterized in that the method comprises:
receiving and responding to a route transmission request, and acquiring a source node and a destination node in the computer network system;
determining an optimal transmission path from the source node to the destination node based on the computer network system; the transmission delay of the optimal transmission path is minimum;
and transmitting data included in the route transmission request from the source node to the destination node based on the optimal transmission path.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311139923.0A CN116886591B (en) | 2023-09-06 | 2023-09-06 | Computer network system and routing method |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311139923.0A CN116886591B (en) | 2023-09-06 | 2023-09-06 | Computer network system and routing method |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN116886591A CN116886591A (en) | 2023-10-13 |
| CN116886591B true CN116886591B (en) | 2023-12-15 |
Family
ID=88262539
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311139923.0A Active CN116886591B (en) | 2023-09-06 | 2023-09-06 | Computer network system and routing method |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116886591B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117610469B (en) * | 2024-01-23 | 2024-05-14 | 芯来智融半导体科技(上海)有限公司 | Topology system based on core particles |
| CN118890306B (en) * | 2024-09-29 | 2025-01-24 | 山东云海国创云计算装备产业创新中心有限公司 | Computer network communication method, computer equipment, storage medium and program product |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015176243A1 (en) * | 2014-05-21 | 2015-11-26 | 华为技术有限公司 | Improved ring topology structure and application method thereof |
| KR101800320B1 (en) * | 2016-06-10 | 2017-11-22 | 성균관대학교산학협력단 | Network on chip system based on bus protocol, design method for the same and computer readable recording medium in which program of the design method is recorded |
| CN108235273A (en) * | 2017-12-30 | 2018-06-29 | 山东省科学院海洋仪器仪表研究所 | A kind of marine mobile sensor network route maintenance method based on collection thimble mechanism |
| WO2019011338A1 (en) * | 2017-07-13 | 2019-01-17 | 华为技术有限公司 | Method for determining shortest path and controller |
| CN111314988A (en) * | 2020-02-20 | 2020-06-19 | 吉林建筑科技学院 | Annular wireless sensor network clustering method based on optimal cluster number |
Family Cites Families (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| EP1748604B1 (en) * | 2005-07-28 | 2010-03-03 | Technische Universität Berlin | A communication network, a method of routing data packets in such communication network and a method of locating and securing data of a desired resource in such communication network |
| US9148298B2 (en) * | 2011-02-03 | 2015-09-29 | Futurewei Technologies, Inc. | Asymmetric ring topology for reduced latency in on-chip ring networks |
| WO2015066365A1 (en) * | 2013-10-30 | 2015-05-07 | Huawei Technologies Co., Ltd. | Network topology of hierarchical ring with gray coding shortcuts |
-
2023
- 2023-09-06 CN CN202311139923.0A patent/CN116886591B/en active Active
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2015176243A1 (en) * | 2014-05-21 | 2015-11-26 | 华为技术有限公司 | Improved ring topology structure and application method thereof |
| KR101800320B1 (en) * | 2016-06-10 | 2017-11-22 | 성균관대학교산학협력단 | Network on chip system based on bus protocol, design method for the same and computer readable recording medium in which program of the design method is recorded |
| WO2019011338A1 (en) * | 2017-07-13 | 2019-01-17 | 华为技术有限公司 | Method for determining shortest path and controller |
| CN108235273A (en) * | 2017-12-30 | 2018-06-29 | 山东省科学院海洋仪器仪表研究所 | A kind of marine mobile sensor network route maintenance method based on collection thimble mechanism |
| CN111314988A (en) * | 2020-02-20 | 2020-06-19 | 吉林建筑科技学院 | Annular wireless sensor network clustering method based on optimal cluster number |
Non-Patent Citations (2)
| Title |
|---|
| 层次环网络的教育资源共享结构及相关策略分析;邢长明;韩庆华;韩勇;;计算机应用研究(07);全文 * |
| 超立方体双环互连网络及路由算法;刘有耀;韩俊刚;;计算机应用研究(03);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116886591A (en) | 2023-10-13 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN116886591B (en) | Computer network system and routing method | |
| US8819611B2 (en) | Asymmetric mesh NoC topologies | |
| JP5276220B2 (en) | Bus control device and control device for outputting instructions to bus control device | |
| CN111786911B (en) | Hybrid wireless optical network-on-chip system and multicast routing method thereof | |
| US10091091B2 (en) | Direct network having plural distributed connections to each resource | |
| US8825986B2 (en) | Switches and a network of switches | |
| TWI661700B (en) | Network topology system and topology building method thereof | |
| Rad et al. | A survey and taxonomy of congestion control mechanisms in wireless network on chip | |
| CN107959643A (en) | A kind of exchange system and its routing algorithm built by exchange chip | |
| US8953497B2 (en) | Modified tree-based multicast routing schema | |
| CN102546417A (en) | Scheduling method of network-on-chip router based on network information | |
| CN116915708A (en) | Method for routing data packets, processor and readable storage medium | |
| Uma et al. | Network-on-chip (noc)-routing techniques: A study and analysis | |
| CN106209518B (en) | One kind being based on the dynamic steering routing algorithm of " packet-circuit " switching technology | |
| Ramakrishnan et al. | Analysis of network on chip topologies | |
| CN103729332A (en) | MoT (mesh-of-tree) structure based low-power-consumption NoC (network-on-chip) routing method | |
| CN115665041B (en) | Network-on-chip structure, data transmission method, electronic device, and storage medium | |
| CN104270308A (en) | A radio frequency network-on-chip application mapping method for unbalanced communication characteristics | |
| CN110825689B (en) | Electronic chip and implementation method thereof | |
| CN105224501A (en) | Improve anchor ring network and determine the method and apparatus in data packet transmission path | |
| CN111817952B (en) | High Fault Tolerance and Low Delay Routing Algorithm Based on Mesh Structure | |
| CN119743450B (en) | Message transmission method and device of network-on-chip, electronic equipment and storage medium | |
| WO2016165072A1 (en) | Method and apparatus for deploying wired return link for wireless access point | |
| CN116028424A (en) | Many-core chip and routing method using same | |
| CN115297065A (en) | Processing equipment communication interconnection method and device, computer equipment and storage medium |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |